Time-Continuous Sliding Motion Image Registration Using Stationary Velocity Fields for Respiratory Motion Interpolation

Ole Gildemeister; Johannes Bostelmann; Pia F Schulz; Andra Oltmann; Phillip Rostalski; Jan Modersitzki; Jan Lellmann

Time-Continuous Sliding Motion Image Registration Using Stationary Velocity Fields for Respiratory Motion Interpolation

Ole Gildemeister, Johannes Bostelmann, Pia F Schulz, Andra Oltmann, Phillip Rostalski, Jan Modersitzki, Jan Lellmann

Institute of Mathematics and Image Computing

Abstract

We propose an image registration model for computing a
dense, piecewise diﬀeomorphic deformation map between 3D thorax
images which incorporates sliding motion as often occurring in the pleu-
ral cavity. Our approach is based on stationary velocity fields and neural
implicit representations, with a particular focus on facilitating motion
interpolation. This allows for generating a time-continuous model of the
respiratory cycle based on end-inspiration and end-expiration images.
We investigate the eﬀect of composing deformations in motion interpo-
lation, using a hybrid technique to enforce domain alignment. We experi-
mentally validate our approach for registration and motion interpolation
between thoracic end-inspiration and end-expiration images.

Original language	Undefined/Unknown
Title of host publication	International Conference on Scale Space and Variational Methods in Computer Vision
Number of pages	13
Publication date	2025
Pages	404-416
Publication status	Published - 2025

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 9 Industry, Innovation, and Infrastructure

Cite this

@inproceedings{a59ca1586ce74e1694a9d9ff0a1c1e21,

title = "Time-Continuous Sliding Motion Image Registration Using Stationary Velocity Fields for Respiratory Motion Interpolation",

abstract = "We propose an image registration model for computing a dense, piecewise diﬀeomorphic deformation map between 3D thorax images which incorporates sliding motion as often occurring in the pleu- ral cavity. Our approach is based on stationary velocity fields and neural implicit representations, with a particular focus on facilitating motion interpolation. This allows for generating a time-continuous model of the respiratory cycle based on end-inspiration and end-expiration images. We investigate the eﬀect of composing deformations in motion interpo- lation, using a hybrid technique to enforce domain alignment. We experi- mentally validate our approach for registration and motion interpolation between thoracic end-inspiration and end-expiration images.",

author = "Ole Gildemeister and Johannes Bostelmann and Schulz, \{Pia F\} and Andra Oltmann and Phillip Rostalski and Jan Modersitzki and Jan Lellmann",

year = "2025",

language = "undefiniert/unbekannt",

pages = "404--416",

booktitle = "International Conference on Scale Space and Variational Methods in Computer Vision",

}

Time-Continuous Sliding Motion Image Registration Using Stationary Velocity Fields for Respiratory Motion Interpolation. / Gildemeister, Ole ; Bostelmann, Johannes; Schulz, Pia F et al.
International Conference on Scale Space and Variational Methods in Computer Vision. 2025. p. 404-416.

Research output: Chapters in Books/Reports/Conference Proceedings › Conference contribution › peer-review

TY - GEN

T1 - Time-Continuous Sliding Motion Image Registration Using Stationary Velocity Fields for Respiratory Motion Interpolation

AU - Gildemeister, Ole

AU - Bostelmann, Johannes

AU - Schulz, Pia F

AU - Oltmann, Andra

AU - Rostalski, Phillip

AU - Modersitzki, Jan

AU - Lellmann, Jan

PY - 2025

Y1 - 2025

N2 - We propose an image registration model for computing a dense, piecewise diﬀeomorphic deformation map between 3D thorax images which incorporates sliding motion as often occurring in the pleu- ral cavity. Our approach is based on stationary velocity fields and neural implicit representations, with a particular focus on facilitating motion interpolation. This allows for generating a time-continuous model of the respiratory cycle based on end-inspiration and end-expiration images. We investigate the eﬀect of composing deformations in motion interpo- lation, using a hybrid technique to enforce domain alignment. We experi- mentally validate our approach for registration and motion interpolation between thoracic end-inspiration and end-expiration images.

AB - We propose an image registration model for computing a dense, piecewise diﬀeomorphic deformation map between 3D thorax images which incorporates sliding motion as often occurring in the pleu- ral cavity. Our approach is based on stationary velocity fields and neural implicit representations, with a particular focus on facilitating motion interpolation. This allows for generating a time-continuous model of the respiratory cycle based on end-inspiration and end-expiration images. We investigate the eﬀect of composing deformations in motion interpo- lation, using a hybrid technique to enforce domain alignment. We experi- mentally validate our approach for registration and motion interpolation between thoracic end-inspiration and end-expiration images.

M3 - Conference contribution

SP - 404

EP - 416

BT - International Conference on Scale Space and Variational Methods in Computer Vision

ER -